Anthranilimide-based glycogen phosphorylase inhibitors for the treatment of type 2 diabetes: 1. Identification of 1-amino-1-cycloalkyl carboxylic acid headgroups

Article abstract of DOI:10.1016/j.bmcl.2008.11.085

Optimization of the amino acid residue within a series of anthranilimide-based glycogen phosphorylase inhibitors is described. These studies culminated in the identification of anthranilimides 16 and 22 which displayed potent in vitro inhibition of GPa in

Full text of DOI:10.1016/j.bmcl.2008.11.085

Bioorganic & Medicinal Chemistry Letters 19 (2009) 976–980
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Anthranilimide-based glycogen phosphorylase inhibitors for the treatment
of type 2 diabetes: 1. Identification of 1-amino-1-cycloalkyl carboxylic acid
headgroups
*
Steven M. Sparks , Pierette Banker, David M. Bickett, H. Luke Carter, Daphne C. Clancy, Scott H. Dickerson,
Kate A. Dwornik, Dulce M. Garrido, Pamela L. Golden, Robert T. Nolte, Andrew J. Peat, Lauren R. Sheckler,
Francis X. Tavares, Stephen A. Thomson, Liping Wang, James E. Weiel
Metabolic Center for Excellence in Drug Discovery, GlaxoSmithKline, PO Box 13398, Research Triangle Park, NC 27705, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Optimization of the amino acid residue within a series of anthranilimide-based glycogen phosphorylase
inhibitors is described. These studies culminated in the identification of anthranilimides 16 and 22 which
displayed potent in vitro inhibition of GPa in addition to reduced inhibition of CYP2C9 and excellent
pharmacokinetic properties.
Received 6 October 2008
Revised 17 November 2008
Accepted 19 November 2008
Available online 27 November 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Glycogen phosphorylase
Type 2 diabetes
Type 2 diabetes is a major worldwide public health problem
with devastating chronic complications such as cardiovascular dis-
ease, retinopathy, neuropathy, and nephropathy. Type 2 diabetes
affects between 6% and 20% of the population of Western industri-
alized societies and the growth rate is predicted to increase at 6%
per year.1 The disease is polygenic in nature and characterized by
hyperglycemia, defects in pancreatic insulin secretion, and insulin
resistance in skeletal muscle, adipose tissue, and liver. In addition
to these derangements, the rate of endogenous glucose production
is significantly elevated in type 2 diabetics relative to healthy sub-
jects.2,3 The liver accounts for approximately 90% of the body’s
endogenous glucose production (the kidney produces the remain-
ing 10%). In normal individuals, hepatic glucose production (HGP)
is tightly regulated by insulin and insulin’s counter-regulatory hor-
mone glucagon. However, in the type 2 diabetic state hepatic insu-
lin resistance coupled with elevated levels of glucagon leads to
excessive HGP which contributes to the observed hyperglycemia.
HGP is the sum of two metabolic processes: glycogenolysis,
which is the release of monomeric glucose from its polymeric stor-
age form called glycogen, and gluconeogenesis, which is the de
novo synthesis of glucose from lactate, amino acids, and glycerol.
Estimates vary on the relative contribution of gluconeogenesis
and glycogenolysis to HGP in humans; however, studies performed
in type 2 diabetics estimate the glycogenolytic contribution any-
where from 12% to 75% of total HGP.3 Despite the discordance
regarding the relative contribution of glycogenolysis to HGP, inhi-
bition of glycogenolysis represents a novel approach to the devel-
opment of new antihyperglycemic agents.
Among potential biochemical targets which could lead to the
amelioration of the hyperglycemic state, glycogen phosphorylase
a (GPa), which catalyzes the phosphorolytic cleavage of the glucose
polymer glycogen at the d-1,4-glycosidic linkage to produce glu-
cose 1-phosphate (G-1-P), has been investigated.4 Herein we report
the optimization of the amino acid fragment of a series of
anthranilimide-based glycogen phosphorylase inhibitors, leading
to potent, glucose-sensitive, orally bioavailable GPa inhibitors.
Compound 1 was identified as a starting point for optimization
following a high throughput synthesis campaign.5 Compound 1 is a
potent inhibitor of human liver GPa (IC50 = 73 nM) and was found
to be 6 times less potent in the absence of glucose, thus reducing
the risk of hypoglycemia during periods of low blood glucose with
this class of inhibitors. While the potency of anthranilimide 1 was
acceptable, the compound was rapidly metabolized both in vitro
and in vivo. In vitro metabolite ID studies identified the cyclohexyl
residue as the major site of metabolism. In addition, analogs clo-
sely related to 1 showed submicromolar inhibition of CYP2C9.
Thus, the primary goals of the optimization were to increase po-
tency, improve metabolic stability, and reduce P450 inhibition
(see Fig. 1).
The general synthesis of the anthranilimides is shown in
* Corresponding author. Tel.: +1 919 483 1104.
E-mail address: steven.m.sparks@gsk.com (S.M. Sparks).
Scheme 1.6 Amide bond formation between the appropriate amino
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.085

S. M. Sparks et al. / Bioorg. Med. Chem. Lett. 19 (2009) 976–980
977
Table 1
In vitro GPa inhibition: phenyl urea substitution
OH
HN
O
O
OH
HN
Cl
NH
O
O
O
N
H
NH
R
O
N
H
1
IC50 (+glu) = 73 nM
IC50 (-glu) = 460 nM
IC50 (cell) =1290 nM
CYP2C9 IC50 = 1000 nM
Compound
R
GPa IC50 (nM)
GPa (cell)a IC50 (nM)
1
2
3
4
5
6
2-Cl-6-Me
2,6-diCl
2,6-diCl-4-OCF3
2,4,6-triCl
2,6-diMe
73
21 (10)
5 (2)
65 (8)
120 (21)
6 (1)
1290 (177)
320
273 (76)
224 (85)
1060 (250)
373 (110)
Figure 1. In vitro profile of anthranilimide 1.
2,4,6-triMe
a
R
O
Values are means of three experiments, standard error is given in parentheses.
HN
CO2Me
R
CO2H
a
H2N
CO2Me
2,6-disubstituted and 2,4,6 trisubstituted phenyl ureas was pre-
pared (Table 1).5 The compounds were evaluated in an enzyme as-
say utilizing human liver glycogen phosphorylase and
subsequently progressed to a cellular assay measuring inhibition
of forskolin induced glycogenolysis.7,8 SAR around the phenyl
group of the urea revealed that 2,4,6-trisubstitution provided the
most potent analogs with the 2,4,6-trimethylphenyl urea deriva-
tive 6 providing the optimal balance of enzyme inhibition (GPa
IC50 = 6 nM) and cellular activity (IC50 = 373 nM).
Having secured that 2,4,6-trisubstition on the phenyl group of
the urea was advantageous for potency, optimization of the amino
acid residue was investigated. Custom and commercially available
amino acid esters were employed directly in the sequence shown
in Scheme 1 to provide the desired anthranilimide targets. The cus-
tom amino esters were prepared via several routes detailed in
Scheme 2.9
Initial investigation of the SAR around the amino acid residue of
the anthranilimides revealed the (S)-cyclohexylglycine residue
provided potent enzyme inhibition and submicromolar cellular
activity but suffered from rapid in vitro and in vivo metabolism.
This observation was reaffirmed with the trimethylphenyl urea
derivative 6 which has excellent enzyme and cellular potency (Ta-
ble 1) but had submicromolar inhibition of CYP2C9 and was rap-
idly metabolized in rat liver microsomes (t1/2 < 15 min).
+
NH2
NH2
I
R
O
HN
CO2H
b, c or c, b
NH
R1
O
N
H
II
Scheme 1. General syntheses of the anthranilimide core. Reagents and conditions:
(a) HATU, i-PrNEt2, DMF, rt; (b) R1PhNCO, pyridine, rt; (c) 2 M LiOH, MeOH, THF.
acid ester and 2-aminonaphthoic acid provided the amide deriva-
tives I, which were subsequently condensed with substituted
phenylisocyanates and saponified to provide the desired anthranil-
imide analogs II. Alternatively, the compounds could be assembled
by reversing the order of steps with saponification of the amide
intermediates I followed by urea formation to deliver the desired
compounds II.
Prior to optimizing the amino acid residue, an initial study of
the SAR around the phenyl urea moiety was undertaken utilizing
commercially available phenylisocyanates. Previous work revealed
that 2,6-disubstitution around the phenyl urea was minimally re-
quired to deliver potent GPa inhibitors and therefore a series of
To further explore the SAR around the cyclohexane ring, an
examination of substitution was undertaken with the goal of
R
R
R
R
b
a
fmoc
fmoc
N
H
CO2Me
CO2H
R
H2N
N
H
CO2H
R
X
X
X
d
c
Cbz
N
CO2Me
O
H2N
CO2Me
H
R
R
R
f
e
H
H
N
H
ClH3N
S
OHC
CO2Me
:
O
CN
Scheme 2. General syntheses of the amino acid esters. Reagents and conditions: (a) TMSCH2N2, DCM; (b) PS-N-methylpiperazine, DCM; (c) N-Cbz-2-(diethoxyphospho-
ryl)glycine methyl ester, DBU, DCM; (d) H2, 10% Pd/C; (f) (S)-+-p-tolyl-sulfoximine, Ti(OEt)4, DCM; Et2AlCN, i-PrOH, THF; (g) HCl, MeOH.

978
S. M. Sparks et al. / Bioorg. Med. Chem. Lett. 19 (2009) 976–980
Table 2 (continued)
Table 2
In vitro GPa inhibition: cyclohexylglycine analogs
a
Compound
R
R0
GPa IC50
(nM)
GPa (cell) IC50
(nM)
CYP2C9 IC50
(nM)
R
HN
O
Me
CO2H
16
Me 10 (3)
350 (60)
9200
R'
R'
NH
O
N
R'
H
a
Values are means of three experiments, standard error is given in parentheses
(nt, not tested).
a
Compound
R
R0
GPa IC50
GPa (cell) IC50
CYP2C9 IC50
(nM)
(nM)
(nM)
increasing the metabolic stability by blocking metabolism at the
cyclohexane ring and decreasing inhibition of CYP2C9 (Table 2).
The effect of the ring size of the amino acid residue was explored
with cyclopentylglycine derivative 7 which maintained enzyme
and cellular potency with a modest decrease in CYP2C9 activity.
Substitution at the 4-position of the cyclohexylglycine residue
was evaluated with difluoro analog 8 and 1,4-disubstituted cyclo-
hexane derivatives 9 and 10, which provided a 2- to 100-fold loss
of enzyme activity with an accompanying increase in 2C9 inhibi-
tion. A greater than 10-fold loss of enzyme activity was seen with
the incorporation of polar functionality exemplified by the pyran
derivative 11, ketone derivatives 12 and 13, and the cis-4-hydrox-
ymethylcyclohexylglycine analog 14, indicating that nonpolar lipo-
6
Me 6 (1)
373 (110)
321 (85)
610
CO2H
7
8
Me 14 (2)
1200
CO2H
F
F
Me 13 (8)
1005 (223)
845 (235)
650
180
philic residues are preferred. Armed with this knowledge,
a-
CO2H
methyl-(S)-cyclohexylglycine derivative 16 was prepared which
maintained activity at the enzyme and cellular level comparable
to cyclohexylglycine derivative 6 but also afforded a >10-fold de-
crease in CYP2C9 inhibition.
To understand the structural requirements for the binding of
the anthranilimide inhibitors to glycogen phosphorylase, X-ray
crystallographic analysis was undertaken. Crystallization studies
Me
9
Cl
Cl
56
H
CO2H
utilizing hLGPa with
a-methyl-(S)-cyclohexylglycine derivative
16 revealed the anthranilimide inhibitors bind to the AMP site of
the enzyme (Fig. 2).10 Interestingly, the amino acid moiety of com-
pound 16 was present in two conformations, A and B, suggesting
10
556
2400
210
H
that cyclic a,a-disubstitued amino acids would be accommodated
CO2H
in the binding site. With this information in hand, a series of ami-
no-cycloalkyl-carboxylic acid (ACxC) derivatives were prepared
(Table 3).
O
O
11
12
13
14
Me 97 (39)
4873 (2170)
>10,000
>10,000
nt
2400
9400
nt
a,a
-Dipropyl derivative 17 was initially prepared and although
less potent at GPa, exhibited dramatically decreased CYP2C9 inhi-
bition. Constraint of the -disubstitution into a ring was exam-
CO2H
a,a
ined with ACxC derivatives 18–23 which revealed that large
lipophilic rings were preferred (21–23), consistent with results
from the cyclohexylglycine analogs. Cyclooctyl analog 22 afforded
excellent enzyme inhibition and cellular activity, as well as a great-
er than 50-fold reduction in CYP2C9 inhibitory activity. Incorpora-
Me 199 (37)
Me 132 (38)
Me 17,000
CO2H
O
H
CO2H
OH
nt
H
CO2H
O
O
15
Me 106 (87)
2,053 (623)
6700
Figure 2.
each of the two hLGPa AMP binding-sites seen in the crystal structure. Each site
contains different major conformer within the site as shown above. Minor
conformers in each site were not built.
a-Methyl cyclohexylglycine analog 16 exists in two conformations in
CO2H
a

S. M. Sparks et al. / Bioorg. Med. Chem. Lett. 19 (2009) 976–980
979
Table 3
Table 4
In vitro GPa inhibition: amino-1-cycloalkyl carboxylic acid analogs
Rat and dog pharmacokinetic data for anthranilimide GPIs
R
Compound
(species)
AUC0?1,po (ng h/
mL)a
Cl (mL/min/
kg)
Vss (L/
kg)
t1/2,
(h)
F
(%)
po
HN
O
16 (ratb)
22 (ratb)
22 (dogc)
1720
4380
620
7.5
1.1
13
0.30
0.28
0.50
4.1
5.4
4.0
37
28
45
NH
O
N
H
a
Dose adjusted oral AUC normalized to 1 mg/kg.
b
c
Sprague–Dawley rats (n = 2), oral dose = 10 mg/kg, IV dose = 2.5 mg/kg.
Beagles (n = 2), oral dose = 5 mg/kg, IV dose = 1 mg/kg.
Compound
R
GPa IC50
GPa (cell)
IC50 (nM)
CYP2C9 inhibition
IC50 (nM)
a
(nM)
With potent GPa inhibitors with reduced CYP2C9 inhibitory
17
18
19
20
732 (59)
720 (117)
279
2517 (863)
9023 (972)
1773 (485)
1390 (63)
>33,000
>33,000
>33,000
>33,000
activity in hand, pharmacokinetic (PK) evaluation was under-
taken (Table 4). Cyclohexylglycine analog 16 was evaluated in
rat PK experiments and AC8C analog 22 was evaluated in both
rat and dog studies. The rat PK profiles of both compounds dem-
onstrated low clearance and volumes of distribution and showed
reasonable exposure upon oral dosing (>25% bioavailability) with
oral half-lives of greater than 4 h. The PK of compound 22 in dog
also exhibited good bioavailability and half-life with slightly in-
creased clearance and lower oral exposure relative to the results
in the rat.
~
CO2H
~
~
~
CO2H
CO2H
CO2H
339 (93)
In summary, the amino acid residue of a series of anthranili-
mide-based glycogen phosphorylase inhibitors was optimized to
improve potency and stability while reducing the potential for
CYP2C9 inhibition. Among the analogs described herein, anthranil-
imides 16 and 22 were chosen for evaluation in pharmacokinetic
studies which supported progression into in vivo efficacy models.
These results, along with additional reports on the SAR and optimi-
zation of this class of glycogen phosphorylase inhibitors, will re-
ported in due course.
21
22
23
36 (5)
5 (1)
585 (76)
498 (86)
443 (84)
4,100
>33,000
nt
~
~
CO2H
CO2H
14 (5)
~
CO2H
References and notes
O
O
1. Zimmet, P.; Alberti, K. G.; Shaw, J. Nature 2001, 414, 782.
2. Lefebvre, P. J.; Scheen, A. J. Eur. J. Clin. Invest. 1999, 29, 1.
24
25
1147
nt
nt
nt
nt
3. (a) Roden, M.; Bernroider, E. Best Pract. Res.. Clin. Endocrinol. Metab. 2003, 17,
365; (b) Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.; Rohde, J. J.; Richards,
S. J.; Souers, A. J.; Szczepankiewicz, B. G. Curr. Med. Chem. 2003, 10, 99; (c)
Staehr, P.; Hother-Nielsen, O.; Beck-Nielsen, H. Diabetes Obes. Metab. 2002, 4,
215.
4. (a) Henke, B. R.; Sparks, S. M. Mini-Rev. Med. Chem. 2006, 6, 845; (b) Treadway, J.
L.; Mendys, P.; Hoover, D. J. Exp. Opin. Invest. Drugs 2001, 10, 439.
5. Evans, K. A.; Li, Y. H.; Coppo, F. T.; Graybill, T. L.; Cichy-Knight, M.; Gale, J.;
Thrall, S. H.; Tew, D.; Tavares, F.; Thomson, S. A.; Weiel, J. E.; Boucheron, J. A.;
Clancy, D. C.; Epperly, A. H.; Golden, P. L. Bioorg. Med. Chem. Lett. 2008, 18, 4068.
6. All novel compounds were characterized by NMR and LC–MS and gave results
consistent with the proposed structures.
~
~
CO2H
CO2H
1391 (139)
970 (308)
91 (16)
CO2Et
N
26
27
>10,000
nt
7. Inhibitors were tested for human liver glycogen phosphorylase enzymatic
~
CO2H
activity using
fluorescence due to product formation was measured on a fluorescence plate
reader (Viewlux, Perkin-Elmer) using 525-nm excitation filter and 595
a fluorescence intensity endpoint assay. The change in
a
emission filter. The hGPa enzyme IC50 values given in Tables 1–4 are average
values of at least 2 replicates where standard deviations are noted, and were
measured in the presence of glucose (10 mM). Due to the specific activity of the
enzyme, a concentration of 10–15 nM glycogen phosphorylase is used in the
assay. Therefore, inhibitors with IC50 determined to be < approximately 5 nM
(Kd < enzyme concentration) cannot be accurately evaluated in this assay
format. Inhibitors falling into this category may have IC50 significantly lower
than the estimate. See Ref. 11 for additional details.
4788 (653)
1070 (302)
4877
~
CO2H
28
27 (5)
nt
8. This full curve assay was designed to detect the inhibition of glycogenolysis
(glycogen breakdown) by test compounds. On the day before the assay the
glycogen in HepG2 cells is prelabeled by overnight inclusion of 14C-glucose in
the culture medium. To begin the assay, the cells are treated with test
compounds, and glycogenolysis is stimulated by forskolin treatment for
60 min. The cells are then lysed and the radiolabeled glycogen in the cells is
~
CO2H
a
Values are means of three experiments, standard deviation is given in paren-
theses (nt, not tested).
quantified. If
a test compound inhibits glycogenolysis, the radiolabeled
glycogen content of the cells will be greater than control (forskolin treated).
tion of heteroatoms into AC6C derivatives 24, 25, and 26 resulted in
diminished inhibitory activity of GPa, consistent with SAR gained
from the cyclohexylglycine series. Indene and tetrahydronaphtha-
lene derivatives 27 and 28 were also prepared and showed good
The hGPa (cell) IC50 values given in Tables 1–4 are average values of at least 2
11
replicates where standard errors are noted. See Ref.
for additional details.
9. Dehydroamino acid synthesis: (a) Schmidt, U.; Lieberknecht, A.; Schanbacher,
U.; Beuttler, T.; Wild, J. Angew. Chem., Int. Ed. Engl. 1982, 21, 776; Asymmetric
Strecker chemistry: (b) Davis, F. A.; Portonovo, P. S.; Reddy, R. E.; Chiu, Y. J. Org.
Chem. 1996, 61, 440; The amino acid for compound 13 was prepared by the
enzyme inhibition but diminished cellular activity (>1 lM).

980
S. M. Sparks et al. / Bioorg. Med. Chem. Lett. 19 (2009) 976–980
0
method of Corey: (c) Corey, E. J.; Noe, M. C.; Xu, F. Tetrahedron Lett. 1998, 39,
5347.
2.57 ÅA at the Advanced Photon Source Beamline 17ID at Argonne National Labs
using an ADSC Q210 detector. The structure was refined to an R-factor of 16.7%
with a Free-R factor of 20.7% with good geometry and was deposited to the PDB
with code 3DD1.
10. Crystallography of compound 16 was carried out as described in Thomson, S.
A.; Banker, P.; Bickett, D. M.; Boucheron, J. A.; Carter, H. L.; Clancy, D. C.;
Cooper, J. P.; Dickerson, S. H.; Garido, D. M.; Nolte, R. T.; Peat, A. J.; Sheckler, L.
R.; Sparks, S. M.; Travares, F.X.; Wang, L.; Wang, T. Y.; Weiel, J. E. Bioorg. Med.
Chem. Lett. 2009, doi:10.1016/j.bmcl.2008.12.085. X-ray data was collected to
11. Evans, K. A.; Cichy-Knight, M.; Coppo, F. T.; Dwornik, K. A.; Gale, J. P.; Garrido,
D. M.; Li, Y. H.; Patel, M.; Tavares, F. X.; Thomson, S. A.; Dickerson, S. H.; Peat, A.
J.; Sparks, S. M.; Banker, P.; Cooper, J. P. WO 2006/052722 A1, 2006.